Oncolytic viruses (OVs) specifically infect, replicate in and kill malignant cells, leaving normal tissues unaffected. Several OVs have reached advanced stages of clinical evaluation for the treatment of various neoplasms (Russell S J. et al., (2012) Nat Biotechnol 30:658-670). Once approved, such viral agents could substitute or combine with standard cancer therapies and allow for reduced toxicity and improved therapeutic efficacy.
In addition to the vesicular stomatitis virus (VSV) (Stojdl D F. et al., (2000) Nat Med 6:821-825; Stojdl D F. et al., (2003) Cancer Cell 4:263-275), other rhabdoviruses displaying oncolytic activity have been described recently (Brun J. et al., (2010) Mol Ther 18: 1440-1449; Mahoney D J. et al., (2011) Cancer Cell 20:443-456). Among them, the non-VSV Maraba virus showed the broadest oncotropism in vitro (WO 2009/016433). A mutant Maraba virus with improved tumor selectivity and reduced virulence in normal cells was engineered. The attenuated strain is a double mutant strain containing both G protein (Q242R) and M protein (L123W) mutations. In vivo, this attenuated strain, called MG1 or Maraba MG1, demonstrated potent anti-tumor activity in xenograft and syngeneic tumor models in mice, with superior therapeutic efficacy than the attenuated VSV, VSV.6M51 (WO 2011/070440).
Data accumulated over the past several years has revealed that anti-tumor efficacy of oncolytic viruses not only depends on their direct oncolysis but may also depend on their ability to stimulate anti-tumor immunity (Bridle S W. et al., (2010) Mol Ther 184:4269-4275). This immune-mediated tumor control seems to play a critical role in the overall efficacy of OV therapy. Indeed, tumor-specific adaptive immune cells can patrol the tissues and destroy tumor cells that have been missed by the OV. Moreover, their memory compartment can prevent tumor recurrence.
Various strategies have been developed to improve OV-induced anti-tumor immunity (Pol J. et al., (2012) Virus Adaptation and Treatment 4:1-21). Some groups have genetically engineered OV expressing immunomostimulatory cytokines. A herpes simplex and a vaccinia virus expressing Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF) have respectively reached phase III and IIB of the clinical evaluation for cancer therapy while a VSV expressing IFN-β has just entered phase I.
Another strategy, defined as an oncolytic vaccine, consists of expressing a tumor antigen from the OV (Russell S J. et al., (2012) Nat Biotechnol 30:658-670). Previously, it has been demonstrated that VSV could also be used as a cancer vaccine vector (Bridle B W. et al., (2010) Mol Ther 184:4269-4275). When applied in a heterologous prime-boost setting to treat a murine melanoma model, a VSV-human dopachrome tautomerase (hDCT) oncolytic vaccine not only induced an increased tumor-specific immunity to DCT but also a concomitant reduction in antiviral adaptive immunity. As a result, the therapeutic efficacy was dramatically improved with an increase of both median and tong term survivals (WO 2010/105347). Although VSV was shown to be effective using hDCT as a tumor associated antigen, there is no way to predict what tumor associated antigens will be effective in a heterologous prime-boost setting.
It is desirable to provide a vaccine vector that can be used to activate the patient's immune system to kill tumor cells with reduced toxicity to normal tissues, for example by activating antibodies and/or lymphocytes against a tumor associated antigen on the tumor. It is desirable if such a vaccine vector displays both oncolytic activity and an ability to boost adaptive cell immunity.